Forests for People

[

] 295

Tropical rainforest monitoring with

ALOS/PALSAR in Brazil and Indonesia

Masatoshi Kamei, Makoto Ono, Tsutomu Yamanokuchi, Nobuhiro Tomiyama and Takashi Ogawa,

Remote Sensing Technology Center of Japan

he Advanced Land Observing Satellite (ALOS) was devel-

oped and launched by the Japan Aerospace Exploration

Agency (JAXA) in January 2006. ALOS carries two optical

sensors and one radar sensor on board: the Panchromatic

Remote-sensing Instrument for Stereo Mapping (PRISM), the

Advanced Visible and Near Infrared Radiometer type 2 (AVNIR-

2) and the Phased Array type L-band Synthetic Aperture Radar

(PALSAR). While optical sensors PRISM and AVNIR-2 cannot

observe the ground surface when it is obscured by cloud, PALSAR

provides 10 m ground resolution fine radar data regardless of

weather conditions. PALSAR also has a wide observation mode

called ScanSAR. The ground resolution of ScanSAR is 100 m, but

it covers a width of 350 km. In addition, PALSAR uses L-band

frequency which is suitable for forest monitoring because L-band

radar waves can penetrate forest leaves and obtain information

on both the tree canopy and below. Therefore, PALSAR data is

attractive to countries which have a large forest area in their terri-

tory – in particular tropical rainforest countries (such as Brazil

and Indonesia) and which have been struggling with cloud cover

when observing forest areas from satellites.

The Remote Sensing Technology Center of Japan

(RESTEC) has supported JAXA with the reception, cali-

bration, validation, analysis and dissemination of ALOS

data since its launch in 2006. RESTEC also supported

JAXA when it developed various data products for

users, particularly in a forest field, such as global mosaic

products. In addition to the collaboration with JAXA,

RESTEC has conducted other forest-related research

and projects by using PALSAR data with domestic and

foreign partners. This article introduces two typical

forest projects by RESTEC in Brazil and Indonesia.

Detecting illegal logging in Brazil

Brazil has a long history of using satellite data for forest

conservation since the 1970s and developed the near

real-time forest monitoring system DETER. This system

contributed to preventing illegal logging, a very serious

threat to forests in Brazil. However, one problem is that

the system uses optical satellite data and cannot observe

the ground surface during almost five months of the

year because of thick cloud. The Brazilian Institute

of Environment and Renewable Natural Resources

(IBAMA) and the Federal Police Department (DPF)

obtained PALSAR data from JAXA, but they did not

have mature analysis techniques.

Thus, based on a request from Brazil, the Japan

International Cooperation Agency (JICA) agreed to

transfer the required techniques to IBAMA and DPF

in 2008. In February 2009, JICA launched a three-year

project called ‘Utilization of ALOS images to support

protection of the Brazilian Amazon forest and combat

against illegal deforestation’ in order to detect and

prevent illegal logging in the Amazon using PALSAR

data. JICA selected RESTEC as an implementing organi-

zation of this project.

During this project, IBAMA and DPF receive PALSAR

ScanSAR strip data approximately every five days from

JAXA. The strip data covers a region 350 km wide and

over 3,000 km long, which means that one single strip

data covers more than 1 million square kilometres.

IBAMA and DPF overlay the strip data on the previous

data observed from about five days before, and then

make a colour composite image using the original soft-

ware developed by RESTEC. The original strip data is

black and white, but the colour composite image helps

users identify land classification and obtain valuable

Illegal logging is a very serious issue in Brazil, with large areas of trees being cut down